A High-Throughput Method for Multiplex Protein Analysis Using LUNARIS™ Biochips and the Biotek 405 TS Microplate Washer High

Application Note

A high-throughput method for multiplex protein analysis using LUNARIS™ BioChips and the BioTek 405 TS Microplate Washer High-quality, fast and semi-automated biomarker quantification for pharmaceutical and clinical applications

Authors: Aniss Jazayeri*, Ruth Steiss**, Valentina Villa*

Sensitivity: Performance: · LoD in single-digit pg/mL range with 3–4 log scale · Within-Run Precision <10% CV and Within-Run Accuracy quantifiable range in 70–130% range · Between-Run Precision <20% CV and Between-Run Accuracy Robustness: in 70–130% range · Uniformity of signal over entire 384-well plate <14% CV

A reliable, easy-to-use, and time-saving protocol. The system is compatible with an automated workflow.

A B

Figure 1: (A) The 405 TM Microplate Washer from BioTek is carrying (B) the LUNARISTM BaseFrame from AYOXXA equipped with 4 x 96-well high-density BioChips.

Abstract

We report the successful adaption of AYOXXA’s LUNARIS™ assay protocol to the 405 TS Microplate Washer, allowing automated plate washing and enabling high-throughput applications. Precise and accurate results were obtained among four 96-well BioChips constituting a 384-well plate. Use of the 405 TS plate washer provides significant advantage over the manual protocol in terms of time and effort. The instrument performs the wash process on an entire plate in less than two minutes and provides consistent results among the four BioChips.

The combination of multiplexed LUNARIS™ assays together with BioTek’s 405 TS Microplate Washer allows the generation of large amounts of sample data with minimal effort and time, important for end users in pharmaceutical and biotechnology com- panies, academic and governmental research institutes, and contract research organizations.

* AYOXXA Bioystems GmbH **BioTek Instruments GmbH

www.ayoxxa.com Introduction LUNARIS™, AYOXXA‘s proprietary beads-on-a-chip multiplex protein platform enables basic to clinical research, across drug development to clinical application. The platform allows simultaneous quantification of multiple protein biomarkers in a format compatible with 96- or 384-well ANSI-SBS* microplates. This plate-based format combined with the modular design of LUNARIS™ BioChips supports testing on 384-well microtiter plates allowing protein analysis of up to 160 samples in duplicates, considering the analysis of standards, blanks and controls.

The 405 TS Microplate Washer includes various features suited to a range of assay applications, from basic ELISA to sensitive cell and bead washing. Its Dual-Action™ manifold allows independent control of aspiration and dispense tubes, while an avai- lable built-in ultrasonic cleaner reduces assay failure, system downtime and maintenance efforts. Automatic buffer switching for up to four wash buffers is optional, and the washer can be controlled using the built-in keypad or via the Liquid Handling Control™ software on a PC.

The standard LUNARISTM immunoassay protocol includes manual washing using a multichannel pipet for convenience, deliver- ing sensitivities in single-digit pg/mL concentrations with a quantifiable range of 3–4 log scales, a coefficient of variation (CV) and recovery rates between 70–130%. However, meeting the requirements of pharmaceutical and clinical applications with respect to robustness and time efficiency using an automated, high-throughput washing procedure is necessary.

Accuracy and precision in microplate washing is essential for assay reproducibility and robustness. Here we describe an adaptation of the LUNARIS™ protocol that employs the 405 TS Microplate Washer (BioTek Instruments, USA) to automate the washing steps and enable high-throughput analysis.

A B C

ASSAY READOUT & ANALYZE REPORT

Figure 2: Schematic of the LUNARIS™ multiplex procedure (A) The LUNARIS™ assay follows the classical sandwich ELISA principle. Samples are added to each well of the BioChip, and targeted analytes are captured by the antibody-coated beads during incubation. The wells are washed, and the captured analytes are labeled with biotinylated detection antibodies and streptavidin- phycoerythrin (SA-PE). The plates are air-dried and subsequently imaged with the LUNARISTM Reader (B) to quantify the fluorescence signal on the beads. (C) The LUNARIS™ Analysis Suite Software provides a tailored report with detailed information on the concentration of target proteins in each sample.

* ANSI-SBS: Microtiterplate footprint dimensions recommended by the Society of Biomolecular Screening www.slas.org

www.ayoxxa.com 2 Material and Methods To evaluate and optimize the washing protocol, the LUNARIS™ Human 11-plex Cytokine Kit was used (catalog number LHCY-20110F). Four LUNARISTM BioChips were assembled to fill up the LUNARISTM BaseFrame. The calibration curves were generated according to the kit handbook. Briefly, seven standards were prepared by serial dilution (1:4) of the Human Cytokine Standard using Assay Diluent 1. The standard contained the following analytes: TNF-α, GM-CSF, IFN-Ɣ, IL-10, IL-12p70, IL-1β, IL-2, IL-4, IL-5, IL-6, IL-8. Three known concentrations (low, medium and high) within the quantifiable range of the cytokine standard curves were spiked into human serum samples (Seralab, UK) and used as controls.

The LUNARISTM BioChips were first washed, and then 5 μL each of standard, blank and spiked control was loaded into the appropriate wells (Figure 3A). After an incubation of 3 hours at room temperature (RT), the plate was washed using the BioTek 405 TS Microplate Washer, as described in the washing instructions (Figure 3A). After washing, 10 μL of detection antibody reagent was added and allowed to incubate for 60 minutes at RT. The plate was washed again, and then 10 μL of SA-PE reagent was added. After a 30-minutes incubation, the final plate wash was performed. The plate was air-dried and imaged using the LUNARISTM Reader (catalog number LAK-001). The quantification of the readout was performed using the LUNARIS™ Analysis Suite Software.

To assess the reproducibility of the wash protocol, the precision and accuracy among the four BioChips constituting the 384- well plate were evaluated. The experiment used a single reagent lot1,2. One BioChip was designated as one run, and Within- and Between-BioChip Precision and Accuracy were defined as Within- and Between-Run (WRP, BRP, WRA and BRA, see» Definitions« on page 6). The WRA and WRP were determined by measuring the controls, 1:2 diluted in Assay Diluent 1. For each control (low, medium and high spike-in concentration), five independent preparations were analyzed on each BioChip, each preparation in triplicate. The BRP and BRA were determined by measuring the controls over the four individual LUNARISTM BioChips.

Automated plate washing was performed using the 405 TS Microplate Washer with a 96-well Dual-Action™ manifold. The specific parameters for each procedure are listed in Figure 3B and have been optimized for washing efficiency and assay precision3.

A B

5 µL Standards, Method Samples, Controls, Wash Buffer 1 A Blanks Wash Buffer 2 B (required in the final wash step) Wash 1 x 50 µL Incubation Plate Type 384 (15 min incubation 3 h at RT Soak / Shake 10 sec before aspiration) Dispense 10 µL Detection Volume 50 µL/well Antibody Flow Rate 3 Incubation Wash 3 x 50 µL Height Z 86 1 h at RT (10 sec shake) Horizontal X 44 Horizontal y 0 10 µL SA-PE Prime Yes Incubation Prime Volume 50 mL Wash 3 x 50 µL 30 min at RT (10 sec shake) Prime Flow Rate 11 Aspiration Reverse Travel Rate 3CW 3 x Wash Buffer 1 Centrifugation Height Z 60 3 x Wash Buffer 2 and dry Horizontal X 0 each 50 µL (10 sec shake) Horizontal y 0 Aspiration Delay 0 Final Aspiration Yes Height Z 38 Horizontal X 5 Horizontal y 5

Figure 3: Automated plate washing protocol and parameters. (A) The LUNARIS™ assay protocol using the 405 TS Microplate Washer. (B) The specific plate washing parameters used in the protocol. www.ayoxxa.com 3 Results and Discussion The wash protocol using the 405 TS Microplate Washer (Figure 3B) was optimized to ensure signal robustness and a desired well-to-well CV below 20%. The precision of the aspirate and dispense manifold was assessed on four 96-well BioChips forming a 384-well plate and by analyzing the measured signal using a constant standard concentration4. The homogeneous signal, measured across the four BioChips, is shown for the representative cytokine IL-10 (Figure 4A).

The inter-replicate CV of each cytokine was calculated per column, per row and across the entire plate (Figure 4B). For all cytokines analyzed, the observed mean inter-replicate CV was around 10% per column and per row. The inter-replicate CV across the plate ranged between 11–14%.

IL-10 Variation A B

30 per Column (n=16) 20 600 10

500 %] 0

V[ 30 400 per Row (n=24) eC 300 20

200 10 replicat 100 P 0 M 30 Plate (n=384) 0 J Inter 20 1 2 3 G 4 5 6 7 8 9 D Median Fluorescence Intensity 10 11 12 10 13 14 15 16 17 18 A 19 20 21 22 23 24 0 γ 2 β 2 4 5 6 8 α N 1 F IF IL- IL- IL- IL- IL- IL-10 IL-1 IL- TN GM-CSF

Figure 4: Variation on four 96-well BioChips (384-well plate) (A) 3-D surface plot of Median Fluorescence Intensity of representative IL-10, measured in each well (n=384 wells). (B) Inter-replicate CV of the signal measured for all cytokines on four 96-well BioChips, per column (24 columns, 16 wells each), per row (16 rows, 24 wells each) and over the entire plate (384 wells), represented in a scatter plot with mean percentage and standard deviation.

Once the automated wash protocol was established, the WRP, WRA, BRP and BRA were obtained by measuring three different concentrations of spiked controls on four 96-well LUNARIS™ Human 11-plex Cytokine Kit BioChips (see Materials and Methods). One BioChip was designated as one run. Therefore, the WRP, BRP, WRA and BRA were defined as within- and between-BioChip. For each BioChip, a single standard curve comprising all cytokines was generated to determine the sensitivities and analytical measurement ranges (Figure 5, Table 1).

A Cytokine B GM-CSF C Performance y y 10000 10000 GM-CSF

GM-CSF nsit nsit

te BioChip1 te IFNγ IFNγ In In IL-10 BioChip2 IL-10 IL-12 IL-12 1000 IL-1β 1000 BioChip3 IL-2 IL-1β IL-4 BioChip4 IL-2 IL-5 IL-4 IL-6 IL-5 100 IL-8 100

Fluorescence IL-6 Fluorescence α TNF IL-8 TNFα 10 10 0.11 10 100 1000 10000 Median Median 0.01 0.11 10 100100010000 0.11 10 100 1000 10000 [pg/ml] [pg/ml] [pg/ml]

Figure 5: LUNARIS™ Human 11-plex Cytokine Kit Performance (A) Standard curves generated by diluting the Human Cytokine Standard 1:4 in Assay Diluent 1. (B) Comparison of GM-CSF standard curve from the four BioChips forming a 384-well plate.β (C) Mean values of limit of detection (LoD left border), lower limit of quantification (LLOQ middle border) and upper limit of quantification (ULOQ right border) obtained from eight individual BioChips, constituting two 384-well plates.

www.ayoxxa.com 4 Table 1: Sensitivity and quantifiable range of the automated washing procedure using the 405 TS Microplate Washer and the LUNARIS™ Human Cytokine 11-plex Kit. Mean, minimum and maximum values of LoD, LLOQ, ULOQ and quantifiable range were calculated from analysis assays run on eight individual BioChips, constituting two 384-well plates, one plate per experimental day.

LoD [pg/mL] LLOQ [pg/mL] ULOQ [pg/mL] Log Quantifiable Range

Analyte Mean Min Max Mean Min Max Mean Min Max Mean Min Max

GM-CSF 0.76 0.20 2.10 1.7 0.4 6.6 1700 1700 1700 3.2 3.2 3.2

IFNγ 6.58 2.40 25.80 12.0 6.8 27.3 6344 1750 7000 3.8 3.2 3.8

IL-10 2.01 0.50 4.70 5.4 0.5 7.8 1625 500 2000 3.2 2.7 3.3

IL-12 4.92 0.01 11.50 10.4 1.2 19.5 4531 1250 5000 3.6 3.1 3.7

IL-1β 0.53 0.10 1.70 1.7 0.3 5.1 1300 1300 1300 3.1 3.1 3.1

IL-2 11.04 5.30 26.60 29.3 11.7 46.9 3000 3000 3000 3.5 3.5 3.5

IL-4 1.24 0.10 3.10 3.2 0.7 11.7 3000 3000 3000 3.5 3.5 3.5

IL-5 2.14 0.40 6.60 5.3 0.9 13.7 3172 875 3500 3.5 2.9 3.5

IL-6 0.61 0.02 1.60 1.4 0.7 2.6 2700 2700 2700 3.4 3.4 3.4

IL-8 0.61 0.10 2.10 1.7 0.4 6.6 1541 425 1700 3.2 2.6 3.2

TNFα 8.94 1.00 31.40 15.0 2.9 46.9 3000 3000 3000 3.5 3.5 3.5

The signals measured for the spiked controls were interpolated in the standard curves to determine WRP and BRP and WRA and BRA, as described above in Material and Methods. For all cytokines and concentrations analyzed, excellent WRP and BRP were observed on each and across the BioChips, with WRP mostly below 10% and BRP mostly below 15% (Figure 6 A–B)2. WRA and BRA was consistently within the acceptance range of 70–130% over four BioChips (Figure 6 C–D).

A B

Within-Run-Precision Between-Run-Precision C1 25 C1 25 C2 C2 20 C3 20 C3 15 15 P% P% 10

BR 10 WR 5 5 0 0 F γ β 2 4 5 6 8 10 12 1 α F γ β 2 4 5 6 8 α FN IL- IL- IL- IL- IL- 10 12 1 L- L- L- L- L- -CS I IL- IL- IL- NF CS IFN L- L- L- I I I I I NF T - I I I T M M G G

C D

Within-Run-Accuracy Between-Run Accuracy C1 200 C2 200 C1 C3 C2 C3 130 130 A% 100 A% 100 WR 70 BR 70

0 0 γ F β 2 4 5 6 8 α γ 10 12 1 L- L- L- L- L- F β 2 4 5 6 8 α CS IFN L- L- L- I I I I I NF 10 12 1 L- L- L- L- L- - I I I T CS IFN L- L- L- I I I I I NF M - I I I T G M G

Figure 6: WRP, BRP, WRA and BRA of the automated washing procedure using the 405 TS Microplate Washer and the LUNARIS™ Human 11-plex Cytokine Kit Three known concentrations within the quantifiable range of the standard curves of all cytokines were spiked into human serum to produce the control samples (low, medium and high spike-in concentration, respectively C1, C2 and C3). (A) and (C) WRP and WRA were determined by measuring the controls from five independent preparations on each BioChip; each prepared control was analyzed in triplicate. (A) Scatter plot shows the WRP obtained per control sample, per BioChip, per analyte. (C) Box plot shows the WRA range measured, with median, min and max, per control, and per analyte. (B) and (D) BRP and BRA were obtained by measuring the controls over four BioChips and in triplicate.

This report summarizes the adaptation of the LUNARISTM assay protocol using the BioTek 405 TS Microplate Washer. The procedure is robust and precise, allows biomarker detection in single-digit pg/mL range over 3-4 log scale quantifiable range. Thus the method is compatible with a semi-automated workflow needed for high-throughput applications. www.ayoxxa.com 5 Definitions

Assay precision Percentage coefficient of variation (%CV) within (WRP) and between analytical runs (BRP). %CV = SD/Mean x 100%, where SD is the standard deviation, and the mean corresponds to the back-calculated concentration of spiked control.

Assay accuracy Percentage recovery to determine the closeness of measurements to the expected concentrations within (WRA) and between the runs (BRA). %Accuracy = Measured concentration/Expected concentration x 100%.

LoD, Limit of Lowest concentration of an analyte for which signals can be distinguished from the blanks Detection (2 x standard deviation of the blank).

LLOQ, Lower limit of Lowest analyte concentrations detectable per analytical assay with the required precision Quantification of ≤20% and accuracy of 70 to 130%.

ULOQ, Upper limit of Highest analyte concentrations detectable per analytical assay with the required precision Quantification of ≤20% and accuracy of 70 to 130%.

References

1 Ong SP, et al. Validation of a Novel Cytokine Microassay for Vitreous Analysis in Diabetic Retinopathy. In review. 2 Committee for Medicinal Products for Human Use (CHMP) (2011). Guideline on bioanalytical method validation. EMEA/CHMP/EWP/192217/2009 Rev.1 Corr.2. 3 Greene J. Maximizing Microplate Washer Performance to Develop Robust and Reproducible Assays (2011) BioTek Webinar, www.biotek.com. 4 Sittampalam GS, Coussens NP, Brimacombe K, et al., editors. Assay Guidance Manual [Internet]. Bethesda (MD): Eli Lilly & Company and the National Center for Advancing Translational Sciences; 2004.

Please note that LUNARISTM products are purely research tools and may not be used for applied diagnostic purposes.

Contact Headquarters GERMANY BioTek Instruments, Inc. AYOXXA Biosystems GmbH 100 Tigan Street BioCampus Cologne Winooski, VT 05404, Nattermannallee 1, 50829 Köln, Germany United States Phone: +49 (0) 221 222 529-0 Phone: +1 617-599 5545 E-Mail: [email protected] E-Mail: [email protected] Web: www.ayoxxa.com Web: www.biotek.com